Copyright © 2006 by the Association for Computing Machinery, Inc. Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, to republish, to post on servers, or to redistribute to lists, requires prior specific permission and/or a fee.

Action recognition is an important and challenging topic in computer vision, with many important applications including video surveillance, automated cinematography and understanding of social interaction. Yet, most current work in gesture or action interpretation remains rooted in view-dependent representations. This paper introduces Motion History Volumes (MHV) as a free-viewpoint representation for human actions in the case of multiple calibrated, and background-subtracted, video cameras. We present algorithms for computing, aligning and comparing MHVs of different actions performed by different people in a variety of viewpoints. Alignment and comparisons are performed efficiently using Fourier transforms in cylindrical coordinates around the vertical axis. Results indicate that this representation can be used to learn and recognize basic human action classes, independently of gender, body size and viewpoint.

future trends

This project represents the results of our ongoing work to efficiently use local shape features and is more thoroughly described in the article “Partial Matching of 3D Shapes with Priority-Driven Search” [Funkhouser and Shilane 2006] with the following abstract. “Priority-driven search is an algorithm for retrieving similar shapes from a large database of 3D objects. Given a query object and a database of target objects, all represented by sets of local 3D shape features, the algorithm produces a ranked list of the c best target objects sorted by how well any subset of k features on the query match features on the target object. To achieve this goal, the system maintains a priority queue of potential sets of feature correspondences (partial matches) sorted by a cost function accounting for both feature dissimilarity and the geometric deformation. Only partial matches that can possibly lead to the best full match are popped off the queue, and thus the system is able to find a provably optimal match while investigating only a small subset of potential matches. New methods based on feature distinction, feature correspondences at multiple scales, and feature difference ranking further improve search time and retrieval performance. In experiments with the Princeton Shape Benchmark, the algorithm provides significantly better classification rates than previously tested shape matching methods while returning the best matches in a few seconds per query.“ Please see the paper for further details. 2

In this paper, we describe a shape-based 3D model retrieval method based on multi-scale local visual features. The features are extracted from 2D range images of the model viewed from uniformly sampled locations on a view sphere. The method is appearance-based, and accepts all the models that can be rendered as a range image. For each range image, a set of 2D multi-scale local visual features is computed by using the Scale Invariant Feature Transform [22] algorithm. To reduce cost of distance computation and feature storage, a set of local features describing a 3D model is integrated into a histogram using the Bag-Of-Features approach. Our experiments using two standard benchmarks, one for articulated shapes and the other for rigid shapes, showed that the methods achieved the performance comparable or superior to some of the most powerful 3D shape retrieval methods. KEYWORDS: Content-based retrieval, multi-scale feature, bag-offeatures,

We consider the use of medial surfaces to represent symmetries of 3-D objects. This allows for a qualitative abstraction based on a directed acyclic graph of components and also a degree of invariance to a variety of transformations including the articulation of parts. We demonstrate the use of this representation for 3-D object model retrieval. Our formulation uses the geometric information

Earlier work by Driscoll and Healy [4] has produced an efficient O(B 2 log 2 B) algorithm for computing the Fourier transform of band-limited functions on the 2-sphere. In this paper, we discuss an implementation of an O(B 4) algorithm for the numerical computation of Fourier transforms of functions defined on the rotation group, SO(3). This compares with the direct O(B 6) approach. The algorithm we implemented is based on the “Separation of Variables ” technique, e.g. as presented by Maslen and Rockmore [19]. In conjunction with the techniques developed in [4], the SO(3) algorithm we implemented may be made truly fast, O(B 3 log 2 B). Basic results will be presented establishing the algorithm’s numerical stability, and examples of applications will be presented. 1

We present an approach to robust shape retrieval from databases containing articulated 3D models. Each shape is represented by the eigenvectors of an appropriately defined affinity matrix, forming a spectral embedding which achieves normalization against rigid-body transformations, uniform scaling, and shape articulation (bending). Retrieval is performed in the spectral domain using global shape descriptors. On the McGill database of articulated 3D shapes, the spectral approach leads to absolute improvement in retrieval performance for both the spherical harmonic and the light field shape descriptors. The best retrieval results are obtained using a simple and novel eigenvalue-based descriptor we propose.

Methods for content-based similarity search are fundamental for managing large multimedia repositories, as they make it possible to conduct queries for similar content, and to organize the repositories into classes of similar objects. 3D objects are an important type of multimedia data with many promising application possibilities. Defining the aspects that constitute the similarity among 3D objects, and designing algorithms that implement such similarity definitions is a difficult problem. Over the last few years, a strong interest in 3D similarity search has arisen, and a growing number of competing algorithms for the retrieval of 3D objects have been proposed. The contributions of this paper are to survey a body of recently proposed methods for 3D similarity search, to organize them along a descriptor extraction process model, and to present an extensive experimental effectiveness and efficiency evaluation of these methods, using several 3D databases.

... of 3D models, the need for retrieval of models has gained prominence in the graphics and vision communities. A variety of methods have been proposed that enable the efficient querying of model repositories for a desired 3D shape. Many of these methods use a 3D model as a query and attempt to retrieve models from the database that have a similar shape. In this paper